Programming three parallel computers

@article{Kallstrom1988ProgrammingTP,
  title={Programming three parallel computers},
  author={Marta Kallstrom and Shreekant S. Thakkar},
  journal={IEEE Software},
  year={1988},
  volume={5},
  pages={11-22}
}
Three commercial systems are considered from a programmer's point of view. The three are the Intel iPSC, a network of Inmos transputers, and the Sequent Balance. The differences in overhead are examined by implementing a solution to the traveling-salesman problem on all three. The evaluation focuses on three major issues in parallel programming: (1) how execution is divided among processing elements and how it is controlled; (2) how data are shared; and (3) how events are synchronized. The… 

Figures from this paper

EMMA2: a high-performance hierarchical multiprocessor
TLDR
The development of a parallel system providing high computational power at a reasonable cost using inexpensive processors is described, and the discussion covers EMMA2's characteristics, architecture, buses, standard modules, I/O, approach to coprocessors, and operating system.
Parallel programming using the MMX operating system and its processor
  • E. Gabber
  • Computer Science
    [1988] Proceedings. The Third Israel Conference on Computer Systems and Software Engineering
  • 1988
TLDR
The author describes MMX design principles, its structure and services, and some MMX timing and speedup measurements of several parallel programs.
VMMP: A Practical Tool for the Development of Portable and Efficient Programs for Multiprocessors
  • E. Gabber
  • Computer Science
    IEEE Trans. Parallel Distributed Syst.
  • 1990
The VMMP (virtual machine for multiprocessors) software package is presented. It provides a coherent set of services for parallel application programs running on diverse multiple input multiple data
Transputer based parallel processing for GIS analysis: problems and potentialities
TLDR
Assessment of the relative merits of different parallel architectures, choice of parallel programming languages and re-design of algorithms to allow effective distribution of the computational and i/o load between individual processors, so performance can be optimized are examined.
Automatic Generation of Parallel Occam Programs for Transputer Rings
Do parallel languages respond to the needs of scientific programmers?
TLDR
The discussion centers on the 'mainline' supercomputers for scientific and engineering applications: vector and scalar MIMD (multiple instruction, multiple data) multiprocessors.
Translating from FP to Occam for systolic algorithms
Implementing Branch-and-Bound Algorithms on a Cluster of Workstations — A Survey, Some New Results and Open Problems
TLDR
It is concluded that given proper tuning a distributed Branch-and-Bound algorithm can yield satisfactory speed-up on a cluster of workstations and be used for combinatorial optimization on multicomputers.
Cache coherence requirements for interprocess rendezvous
TLDR
This paper examines the requirements for cache coherence protocols to support efficient interprocessor rendezvous and shows that invalidation protocols are found to be more efficient when processes migrate freely as a result of the scheduling strategy.
...
1
2
3
...

References

SHOWING 1-10 OF 12 REFERENCES
Exploiting multiprocessors: issues and options
TLDR
This paper examines problems involved in moving applications programs to multiprocessors and outlines strategies currently being considered to resolve these problems, finding that there do not currently have any easy solutions.
Concepts and Notations for Concurrent Programming
TLDR
This paper identifies the major concepts and describes some of the more important language notations for writing concurrent programs and three general classes of concurrent programming languages are identified and compared.
Occam and the transputer
  • D. May, R. Shepherd
  • Computer Science
    European Workshop on Applications and Theory in Petri Nets
  • 1988
This paper examines the design of the distributed programming language occam and its implementation on the transputer, a VLSI microcomputer.
Communicating sequential processes
This paper suggests that input and output are basic primitives of programming and that parallel composition of communicating sequential processes is a fundamental program structuring method. When
A Tutorial Introduction To Occam Programming
TLDR
This presentation explains the development of the Occam 2 language definition and some of the techniques used to develop and implement it.
A Parallel Program Event Monitor
  • Proc. Hawaii Int'lConf. System Sciences
  • 1987
Balance : A Shared - Memory Multiprocessor System
  • The Traveling - Salesman Problem on a Hyper - cubic , MlMD Computer
  • 1985
Concurrent Processing: ANew Direction in Scientific Computing
  • Proc. Nat'I Computer Conf
  • 1985
...
1
2
...